Combined detection of COMP and CS846 biomarkers in experimental rat osteoarthritis: a potential approach for assessment and diagnosis of osteoarthritis

Osteoarthritis (OA) is a progressive degenerative joint condition that is estimated to affect over 250 million individuals worldwide [1]. Currently, there is no effective cure to treat OA [2]. Presently, the diagnosis of OA is based on clinical symptoms and traditional radiology. Magnetic resonance imaging (MRI) not only assesses changes in cartilage during OA development, but also assesses joint tissue damage. However, due to financial constraints and the lack of an internationally validated assessment scale, the use of MRI in the routine diagnosis of OA is limited [3], and X-rays remain the routine diagnostic method. Although the accuracy and sensitivity of these approaches are relatively high, these methods fail to distinctively identify the developmental stages of OA. Similarly, most of the current OA treatments are largely palliative until the articular cartilage has been severely damaged; during the development of OA, the joints become completely dysfunctional and prosthetic replacement becomes necessary [4]. Therefore, an effective method for early diagnosis of OA is imperative, and early changes in OA may be reversed by effective therapeutic drugs.

OA is often not detected until the middle or end stage [5‐7]. In recent years, OA biomarkers have received increased attention as an objective indicator for early diagnosis of OA and for assessment of the disease process [8]. OA biomarkers can be detected at an early stage of OA, prior to the presence of radiographic signs. Cartilage oligomeric matrix protein (COMP) is a non-collagenous component of cartilage, which accounts for roughly 1% of the wet weight of articular tissue [9]. COMP promotes the combination of type II collagen fibers and stable fiber networks. Chondroitin sulfate (CS) is a glycosaminoglycan (GAG) that is covalently attached to specific proteins to form proteoglycans, which are abundant components of the extracellular matrix (ECM) [10]. CS 846 epitope (CS846) is a CS synthetic marker and inseparable from the degree of joint injury in patients with OA. CS846 is a by-product of proteoglycan metabolism and has been detected in serum and synovial fluid (SF) [11]. Therefore, CS846 and COMP as sensitive and specific biomarkers reflecting degradation of cartilage and synovial tissues have increasingly been used as helpful promising tools in early OA diagnosis before irreversible damage has occurred.

However, in many studies of early OA diagnosis, only one biomarker in the pathological process of OA was measured at a single time point [12, 13]. In verifying our hypothesis that COMP and CS846 can be used as a joint biomarker for knee OA, a longitudinal study of the combined measurement of serum levels of COMP and CS846 was performed using an anterior cruciate ligament transection (ACLT) rat models in which the relationship between COMP and CS846 levels and the degradation of articular cartilage were evaluated. The area under the curve (AUC) of COMP, CS846, and combined biomarkers in the evaluation and diagnosis of osteoarthritis was analyzed by ROC curve analysis. When promising, combined biomarkers that were identified for clinical use may have a beneficial effect on both patient health and the medical economy.

In this study, we combined the expression of serum levels of COMP and CS846 in OA rats to provide a reference for early diagnostic criteria of OA. A sensitive and stable biochemical indicator was established. In the past few years, the importance of serum biomarkers in OA, synovitis, joint effusion, and the progression of cartilage damage has become increasingly appreciated and has been proven helpful for monitoring disease progression and clinical therapy [1, 17]. However, studies focusing on the magnitude and timing of the effects of OA on serum marker levels are limited. To our knowledge, this is the first study in which the value of combined detection of serum levels of CS846 and COMP was evaluated in early OA diagnosis. In addition, the ACLT model, in which permanent instability in the articular cartilage knee joint occurs, is a frequently used and well-accepted OA model [15, 18]. Thus, the rat OA model used in this study is an excellent translational model for estimating intra-articular inflammation with outcomes that are relevant to human OA [19].

Articular cartilage is composed of hyaline cartilage, which is composed of an abundant ECM and chondrocytes embedded therein, and with mechanical properties [2]. Degradation of the network of collagen and proteoglycans in OA cartilage leads to a loss in tensile strength and shear properties of cartilage [20]. COMP is a tissue-specific protein that binds to the type II collagen network. When articular cartilage is destroyed, COMP is released into the SF and subsequently absorbed in the serum [21]. COMP has shown promising effects as a prognosticator of rapid joint destruction in humans with levels in both serum and SF being increased in patients with OA when compared to those without joint disease [22]. Interestingly, Geng et al. reported that COMP deficiency enhanced the early onset and development of chronic arthritis but does not affect collagen II autoimmunity [23]. In addition, in a study performed by Fernandes et al., it was suggested that COMP could be used as a diagnostic marker for early OA as shown by elevated serum COMP levels in patients with knee pain symptoms without any radiological abnormalities, thereby indicating early cartilage damage in patients when compared to healthy controls [24]. Moreover, in a study by Huebner et al., SF levels of keratin sulfate (KS) and COMP increased, which was coincident with histological OA, and correlated positively with the severity of histological damage in guinea pig [25]. In our study, we showed that at 2 weeks after ACLT surgery, rats in the model group had significantly higher serum levels of COMP when compared to rats in the control group that at 4 weeks after surgery, they started to rapidly increase. In addition, at other time points, serum COMP levels of rats in the OA group were also higher when compared to those in the control group. In a previous study, similar results were obtained in rabbits with ACLT-induced OA [26]. These results indicated that serum COMP levels in rats in the model group increased early in OA development and gradually increased within 10 weeks after ACLT surgery. The increased release of COMP into serum during early OA is likely due to series of catabolic events that occur in articular cartilage, resulting in a high turnover rate by chondrocytes to repair the cartilage matrix. This process first led to dismantling of the cartilage matrix and subsequently a net loss of tissue. This inference is supported by our histological results, which indicated minor damage in the articular cartilage in rats in the model group at week 2 after surgery. Similar results were obtained by Bin et al. [27]. Also, the increase in SF COMP levels correlated with the length of time post injury, suggesting that COMP may be a suitable marker for longitudinal studies to evaluate its role in joint healing [28]. Georgiev et al. observed higher serum COMP levels in knee OA patients when compared to controls, and COMP correlated positively with Whole-Organ MRI Score, suggesting that COMP may reflect structural damage of the knee joint [29].

CS846 is a synthetic CS marker and inseparable from the degree of joint injury in patients with OA [10]. CS846 levels were found to be significantly increased in serum and SF of experimental equine OA models when compared to healthy controls, and levels further increased in response to exercise [30]. Levels of CS and CS846 were significantly increased in patients with juvenile idiopathic arthritis [31]. Previous studies have indicated that CS846 was almost absent from mature adult cartilage; however, CS846 is present at increased levels following OA [32, 33] and joint injury [34]. In a previous study, Svoboda et al. reported that changes in serum levels of CS846 from baseline to follow-up were significantly different between ACL-injured patients and uninjured controls [35]. In our study, we found that at all time points after surgery, serum levels of CS846 in rats in the model group were higher when compared to those in the control group. In addition, we found that CS846 serum levels increased significantly at 2 weeks after ACLT surgery. This may explain the degradation of proteoglycans observed in the early onset of OA. These findings were supported by our macroscopic observation and histological results, indicating that there was minor damage of articular cartilage in the model group at 2 weeks after surgery. Nevertheless, in the present study, post injury changes in CS846 between ACL-injured cases and uninjured controls were comparable with those observed between OA patients and healthy controls as reported by Svoboda et al. [35]. In previous studies, it has been suggested that the combination of urinary CTX-II, serum COMP, and serum CS846 levels is indicative of the amount of joint damage in patients with hemophilic arthropathy [36]. Based on the data obtained from prior studies and ours, we hypothesized that CS846 can be used as an OA biomarker to assess the extent of articular cartilage damage. In addition, we found that serum levels of CS846 and COMP increased with the development of OA, and the peak value was not detected at 10 weeks after ACLT surgery.

As the sensitivity of a single serum OA biomarkers in predicting OA was low, thereby limiting its potential in clinical application. Therefore, we analyzed the sensitivity when these markers were combined and obtained the AUC of ROC curve. We then calculated their diagnostic value in OA. The combined biomarkers yielded a ROC value of 0.926, which was significantly higher when compared to that of the single biomarker and can be better used for evaluation and diagnosis of OA. In previous studies, early detection of OA has been reported by a combination of biomarkers [26, 27, 37]. Nowadays, the search for early diagnostic markers of OA has become a hot topic. To investigate the correlation between levels of COMP and CS846 and the severity of articular cartilage degeneration, histological changes in articular cartilage at various time points were evaluated based on the OARSI scoring system. In this study, the OARSI scores of rats in the model group were significantly higher regarding both OARSI scores and macroscopic observation scores when compared to the control group. There was a highly positive correlation between COMP and CS846 levels and the OARSI scores (r = 0.915, r = 0.912, P < 0.001). We found that the higher the levels of COMP and CS846 in the serum of ACLT rats, the more severe the degree of articular cartilage lesions. The correlation of COMP was higher compared to that of CS846. Histological examination showed that within 10 weeks after the induction of OA, the surface roughness of articular cartilage gradually increased and the chondrocytes first became hypertrophied and then decreased. In addition, chondrocyte disorganization gradually increased, the intensity of safranin O staining decreased, and the matrix gradually degraded. At 10 weeks after surgery, the stroma showed almost no staining, and clear dissociation of chondrocytes was observed in the lesion area. Histology showed that at 2 weeks after ACLT surgery, the joint surface was rough and after 4 weeks, the degree of joint unevenness was aggravated. Similarly, macroscopic observations also demonstrated that the articular surface was rough at 2 weeks after surgery, and depression appeared on the articular surface at 4 weeks after ACLT surgery, followed by gradual increase of joint surface ulceration and destruction. In the ACLT model used in this study, articular cartilage degeneration was a slow and gradual process, which was in accordance with the true histological processes of OA. These findings indicated that the OA model established in this study was successful. Moreover, we found a positive correlation between changes in levels of COMP and CS846 in rat serum over the course of 10 weeks. As a biomarker, the serum concentration of COMP can reflect the degree of denaturation of type II collagen, whereas serum levels of CS846 can reflect the degree of proteoglycan degradation [35]. Together, these results showed that the degradation of both collagen and proteoglycans resulted in loss of the ECM, thereby causing progressive degeneration of articular cartilage. Synovial hyperplasia, fibrosis, secretion of inflammatory mediators, and cartilage degrading enzymes can lead to acceleration of OA, which intensifies the degradation of proteoglycans and the presence of aggrecan fragments bearing the CS846 epitope. The destruction of cartilage further degrades the collagen network and exacerbates the release of COMP into the SF, which is absorbed by the blood. ECM degradation of chondrocytes is a result of the combined action of type II collagen and proteoglycans [38]. The combined detection of COMP and CS846 fully reflects the degeneration of cartilage. This also indicated that the two indicators of early lesions in response to OA can be used as joint biomarkers for the diagnosis of OA. These results supported the idea that a combination of biomarkers may relate significantly better to the severity of joint damage than individual biomarkers do.

A single test for CS846 or COMP only partially reflected ECM degradation. Combined detection of serum CS846 and COMP concentrations predicted overall degradation of type II collagen and proteoglycans, thereby assessing the degree of OA and early diagnosis of OA. Although the results are promising, our study also has some limitations. Both macroscopic observational histological tests at 10 weeks after OA induction in rats in the model group showed that the degree of OA was significant and that the longitudinal study of 10 weeks was not sufficient. Accordingly, the peak value of CS846 and COMP levels in serum could not be determined at the end of this study. We only performed histological analysis of the tibial plateau, not including the femoral condyle. However, the results were reliable because OA caused articular cartilage degradation in the tibial plateau within 10 weeks after ACLT surgery.

This study provides valuable information on the relationship between serum levels of COMP and CS846, and the degree of OA. Within 10 weeks after ACLT, levels of COMP and CS846 increased significantly, and a positive correlation was observed between the two biomarkers. Serum levels of COMP and CS846 positively correlated with the OARSI score during the OA progression. The combined biomarkers yielded a ROC value, which was significantly higher when compared to that of the single biomarker and would be better if used for the diagnosis of OA. In conclusion, combined detection of serum levels of COMP and CS846 has potential in assessing the degree of OA and OA diagnosis.